What are the hardware requirements for Llava V1.5 13B?

Llava V1.5 13B requires approximately 26 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Llava V1.5 13B and how large is it?

Llava V1.5 13B is developed by liuhaotian, a model with 13b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Llava V1.5 13B?

Llava V1.5 13B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Llava V1.5 13B by liuhaotian — VRAM 26GB, 4K context

Name: Llava V1.5 13B
Author: liuhaotian

Llava V1.5 13B is an open-source language model by liuhaotian. Features: 13b LLM, VRAM: 26GB, Context: 4K, LLM Explorer Score: 0.14.

Image-text-to-text Llava Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/liuhaotian/llava-v1.5-13b

Llava V1.5 13B Parameters and Internals

Model Type

multimodal, chatbot, transformer, auto-regressive

Use Cases

Primary Use Cases:

Research on large multimodal models and chatbots

Training Details

Data Sources:

LAION/CC/SBU (558K filtered image-text pairs), 158K GPT-generated multimodal instruction-following data, 450K academic-task-oriented VQA data mixture, 40K ShareGPT data

Methodology:

Fine-tuning LLaMA/Vicuna on GPT-generated multimodal instruction-following data

Model Architecture:

Transformer

LLM Name	Llava V1.5 13B
Repository 🤗	https://huggingface.co/liuhaotian/llava-v1.5-13b
Model Size	13b
Required VRAM	26 GB
Updated	2026-06-30
Maintainer	liuhaotian
Model Type	llava
Model Files	0.1 GB 9.9 GB: 1-of-3 9.9 GB: 2-of-3 6.2 GB: 3-of-3
Model Architecture	LlavaLlamaForCausalLM
Context Length	4096
Model Max Length	4096
Transformers Version	4.31.0
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32000
Torch Data Type	float16

Rank the Llava V1.5 13B capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Quantized Models of the Llava V1.5 13B

Model	Likes	Downloads	VRAM
SpaceLLaVA	26	233	26 GB
Llava V1.5 13B GPTQ	39	103	7 GB
Llava V1.5 13B AWQ	36	88	7 GB

Best Alternatives to Llava V1.5 13B

Best Alternatives	Context / RAM	Downloads	Likes
Llava V1.6 Vicuna 13B	4K / 26.7 GB	70586	61
Evalalign V1.0 13B	4K / 26.7 GB	3	1
Table Llava V1.5 13B	4K / 26.6 GB	2	1
Posterllava V0	4K / 26.6 GB	1298	7
Peenoblanc 52 Llava	4K / 26.7 GB	6	0
Otterchecker	4K / 26.7 GB	1	2
Llava 13B Custom	4K / 26.7 GB	5	0
V Mllm 13B	4K / 26.6 GB	7	1
...roduct Taxonomy Llava V1.6 13B	4K / 26.7 GB	7	0
Som Llava V1.5 13B	4K / 26.7 GB	12	5

Note: green Score (e.g. "73.2") means that the model is better than liuhaotian/llava-v1.5-13b.